vocaloidfandomcom-20200222-history
Talk:MEIKO/@comment-217.28.13.13-20180305161159/@comment-53539-20180305172443
Actually there are a few things #Meiko is the 4th vocaloid and came after Miriam, Lola and Leon so some bugs were already fixed by her. The only flaws would be related to the Japanese voicebank. #V1 was robotic and unclear overall, its not her all 5 are the same. Some producers just got good at Vocaloid1 usage. #Compactiablity of vocal types also effects Vocaloid and Kaito was a lot more compactable then Meiko. #Kaito didn't have the same version of V1,he had a slightly updated version, 1.1, which meant he was by default better sounding. #The V1 vocal is actually inferior to the V3 version overall, a lot of people claim the V1 is superior but this is not the case. The V1 vocal had a lot of "give" due to V1's crudeness and this was also noted by Wat of CFM. It let him get away with a lot of things he can't in V3. In fact, his V3 vocal actually has a bigger vocal range which is often claimed otherwise. V1 didn't allow for large vocal ranges at all, but did make it appear they had bigger ranges as it was based on an analytical system. Though it had samples, those samples were used for reference to warp the synthiser and were not the primarily focus. We do not hear much of the Kaito voice at all, and V1 was actually hitting sounds much harder and harsher then V1. So even though you may think V3 is the one hitting harder and harsher, he is not. His samples for V3 were made to match V1, meaning that if you hear the raws, not the voice inside the engine but the samples themselves, they sound appox. the same. It is not possible for a 100% match and Vocaloid itself has some impact on the voice. Technology means modern samples can be recorded in a high quality then what was offered in the early 2000s. V3 was recorded, for example, with a digital mic, while the V1 was recorded with the previous generation of mics (forget what they were called). Also, Yamaha handled Kaito V1 while Crypton handled V3. A change of studio also results in some differences due to practices. Even if you still think after that V3 is worst then V1... Consider that V1 lacked things like triphones and was recorded with a worst mic, the quality alone makes it worst if no other things are taken into account. Edit: Bare in mind the average fan doesn't actually know much about these things unless its opening stated. I personally quote the issues with the VY's, that people were told they were HQ so focused on this trait alone. The problem was they only knew they were HQ because they were told. The same can be for Kagamine Len, who is regarded by Act1 as LQ along with Rin. However, the average fan who states things like "Len is the best vocaloid vocal!" and so fourth were largely ignorant to the issues with his vocal. This made it embarrassing to tell people the reality. Meanwhile in another instance Gumi was often quoted as one of the most realistic Vocaloids in V2 in 2010 before V3 hit the table, when she had lost it to miki in late 2009. And lets not mention English vocaloids who were often rated LQ or "less realistic" despite this not being the case. In fact Miku, Len, Rin and even Luka were based on vocal performance and not vocal realism and were far less realistic then the average English vocaloid. And Luka was the HQ Japanese of the non-append CFM vocaloids, but still bottom of the top (which is not a bad position to be frank as only Iroha, VY1, VY2 and the Appends are above her). I've heard somany people presume because they like something or hate something that a voicebank is better or worst without knowing the technicalities or overall quality that it has become a clause of concern, just as my concern for the obsession for clarity above all other traits, or at the cost of them, is a concern to me.